"A box later" Dumble

[WhopperPlate]

I don't believe Dumble ever sold anything to Mayer. I recall interviews of associates of Dumble alluding that Dumble strongly disliked Mayer and his misogynistic ways (guitar wanker interview, I think). https://www.guitarwank.com/podcast/epis ... -15th-2022

The story is Josh @ JHS cloned (or built something inspired by) the "a box later" and called "Box it later" in a smaller footprint. Naturally, this is my entry point to learning more about this little device. I did reach out to Josh but nothing substantial in his reply to assist here.

OK, I must be misremembering how JM acquired this piece. I thought HAD blacklisted him after JM let the two rock guys inside SSS 002 and they cloned it as the Oak Cliff Special. But I imagine it could be for all of the above...

-Dan

Mayer probably bought it used because…like everything else he owns …

[rccolgan]

Like Dan, I found the schematic for a Marshall that seems to be a good candidate. Change the 47nf coupling caps to 1uf and I think we're onto something. Especially if that one large axial is 47uf. One thing that is throwing me off is what looks to be a blue tantalum capacitor on the signal board. It's the same blue tantalum cap that I found he used when I was looking through my Dumble amp pics. The other thing that is throwing me off is what looks like a diode (Edit. Looks like a dark resistor now.. perhaps carbon comp)

[GPD]

Got a lot of "Popular Electronics" to read thru from the 60s & 70s

https://deramp.com/swtpc.com/PopularElectronics/

Also looking up Barcus Berry schematics but I don't think I've come across anything good so far. That 2k pot is throwing me off as a usable level pot. Lowest I've seen is 10k in that spot. Funny enough, that is with the Vemuram TSV808 clone I did.

I'm not certain why a 2K pot value for Drive control would throw you off. Drive = Output and since the desire is for a low impedance output the 2K value is very appropriate. Granted, you might not have seen this particular circuit before but I would not say the 2K drive control pot value is out of the ordinary and pretty typical for a solid-state, OpAmp circuit...maybe not any commercially available solid-state effects loops but surely not out of the ordinary.

[rccolgan]

Got a lot of "Popular Electronics" to read thru from the 60s & 70s

https://deramp.com/swtpc.com/PopularElectronics/

Also looking up Barcus Berry schematics but I don't think I've come across anything good so far. That 2k pot is throwing me off as a usable level pot. Lowest I've seen is 10k in that spot. Funny enough, that is with the Vemuram TSV808 clone I did.

I'm not certain why a 2K pot value for Drive control would throw you off. Drive = Output and since the desire is for a low impedance output the 2K value is very appropriate. Granted, you might not have seen this particular circuit before but I would not say the 2K drive control pot value is out of the ordinary and pretty typical for a solid-state, OpAmp circuit...maybe not any commercially available solid-state effects loops but surely not out of the ordinary.

Thank you for your contribution. My world for a bit has been with pedals where the output impedance resistor is usually 100k or so. Studying the different designs in FX loops is new to me so I'm happy to learn from this project and contributions like yourself. I wish I paid more attention in Op Amp class in college! ha!

Anyhoooo, I added some working files on GitHub if you like to use my working files to move things along. https://github.com/colganr/pedal-wiring ... R_RESEARCH I started a DIYLC layout as well as a photoshop visual to aid in documenting the circuit. I'd like to add a best-guess schematic in the white space once I do some more reading... I will say documenting the board in DIYLC was very helpful in clearing up the low-resolution photos. "Once in a while you get shown the light. In the strangest of places if you look at it right"

Happy to hear input from others here... or a clear gut shot so I can enjoy some time watching TV

[rccolgan]

Alright.. This probably doesn't resemble the Dumble box at all, but I think it covers all the possibilities of using one circuit board to build my artistic license device, as well as the real circuit should it ever be revealed....

A few things to note,

• near the input and output, those are internal trim pots. Input impedance is usually up to 1M on pedals but on the Marshall schematic, it can go as low as 68k. This trim pot accommodates both. Similarly, with the output trim pot, the Marshall output resistor on the Marshall 2203 FX loop is 680k. Usually, guitar pedals are 100k fixed resistors or 10k level pots depending on the pedal.

• I added the 100R and 100pf RF blockers. They can be jumped if you want to but they are becoming commonplace for pedals. I think it makes sense to use them here for long cable going into this device potentially.

• the second op-amp stage, I added a flat boost circuit that goes from 0db to 24db. On the Dumble, if that blue tantalum cap is indeed a thing, there likely is some sort of boost manipulation going on that fits the bill in the design. I just don't know what it is so this should cover it.

• I'm not sure what the return pot value is but I think 1M should cover all scenarios since it's coming in. You'll see I added the same RF blocker 100R/100pF combo on the return.

• The send/return will be connected if no jacks are added (not seen on this schematic, but with the offboard wiring). Using a footswitch, this circuit could accommodate a flat boost or buffer and/or true bypass

Questions I still have

• The 4x film caps... I am unsure what the second 1uf? is for.

• The two sets of orange caps are on-board power supply filters, I'm guessing since it's best practice to have power filter caps as close to the op-amp as possible.

That's about all my energy towards this thing at the moment (that's what I say, but I'll probably be up all night designing the board haha).

Thoughts?

[rccolgan]

I know this thread kinda came to a hault with any Dumble circuit details besides a few educated guesses based on the internals. I'm moving forward with an experiment of mine to hopefully be within striking distance of the original Dumble with internal impedance trim pots and circuit topology that is general enough to substitute components here and there should more info on the Dumble be brought into light.

1590BB or 1590BBS compatible circuit board. The footswitch includes the 9v to +/-15vdc DC-DC converter circuit and there is a dedicated 4 layer board for just the signal components. The DC-DC converter operates at 100khz so well above audible range.

I bet it'd work in this Hammond enclosure too. https://www.amplifiedparts.com/products ... 4-x-55-x-2

Photo album of my notebook and build once the parts come in: https://photos.app.goo.gl/zfz8smN7v1KggQku9

[rogb]

A fascinating project! One point I've thought about, would an increase from 9v to say 25vDC with a MAX1044 charge pump help with headroom, as the K-lator uses ~36vDC? Is not 15v a little low?
Here's a veroboard from tagboardeffects (an RG Keen design, I believe

25V Charge Pump.png

with 25vDC.

I've got a tube Dlator and it's really great in my rack rig. But I'd like to build the Box and a K-lator too, I suppose, just to see if there is any real difference between the 3 approaches.

I built a splitter/buffer for my wet/dry rig recently with 2x 2N5457 JFETs and an EHX LPB1 with a spare 2N5088, which actually works really well. So I'm really into these kind of circuits, hence my interest.

At the moment, I'm near a really great parts supplier in Helsinki, Uraltone, that you can actually buy bits over the counter, like the old days! . Before I go home, I'd like to grab a good selection of parts for the Box and K-lator.

I really appreciate the effort you've made, RC, pushing this project forward.

[dbharris]

I think that would work just fine. You would need to add a voltage divider and filtering for vref at half the voltage output.

Ryan's supply is +/- 15v bipolar, so it is the equivalent of a +30v design in terms of voltage headroom.

-Dan

[rogb]

I think that would work just fine. You would need to add a voltage divider and filtering for vref at half the voltage output.

Ryan's supply is +/- 15v bipolar, so it is the equivalent of a +30v design in terms of voltage headroom.

-Dan

Thanks, you are quite right, it's not a unipolar PS... the +/-15v swing would give 30v.

By the way, has anyone got a Kleinulator schematic, so I can compare the circuits. I found one with series/parallel circuitry, but would be interested in seeing the standard 36v version.

[rootz]

Ryan, nice work as usual. I think your DIYLC layout was a lot closer to the actual loop though.

I have got an idea what the 1uF box caps could be. Might be part of the power supply on the op amp pcb. Look at how symmetric the outsides of that pcb look. Dual electrolytic caps, resistor, 1 uF box cap. Maybe some kind of CRC filter with an extra 1uF cap? still doesn't make much sense to me, as the supply is already regulated?

Then there is that tant. Is that in the ground connection of the feedback loop of the recovery amp? And why, if there is already a symmetrical power supply and output offset can be dialed close to zero out?
And if those 100n caps and the 47uF to the 2k pot are all in the signal line, wouldn't that mean one end of the op amps is still "open" to DC offset? So maybe no input cap from the input of the Box Later (so from the preamp out), but coupling caps to and from effects and to the power amp?

In any way, I've got way more questions than answers...

[rccolgan]

Ryan, nice work as usual. I think your DIYLC layout was a lot closer to the actual loop though.

I have got an idea what the 1uF box caps could be. Might be part of the power supply on the op amp pcb. Look at how symmetric the outsides of that pcb look. Dual electrolytic caps, resistor, 1 uF box cap. Maybe some kind of CRC filter with an extra 1uF cap? still doesn't make much sense to me, as the supply is already regulated?

Then there is that tant. Is that in the ground connection of the feedback loop of the recovery amp? And why, if there is already a symmetrical power supply and output offset can be dialed close to zero out?
And if those 100n caps and the 47uF to the 2k pot are all in the signal line, wouldn't that mean one end of the op amps is still "open" to DC offset? So maybe no input cap from the input of the Box Later (so from the preamp out), but coupling caps to and from effects and to the power amp?

In any way, I've got way more questions than answers...

Interesting! Yes, I agree that the 1uF boxes could be for the power supply. Definitely overkill for such short wires to the op amp. I would think that the second end of the op amp would be part of the recovery part of the circuit.

Yeah the DIYLC layout was just a stab at documenting what was in the real A Box Later and counting parts. I'm with you about having more questions than answers at the moment.

Meanwhile, I have my circuit boards arriving this Wednesday so I can build S/N 0001 of my interpretation. Mouser & StompBoxParts parts have arrived. I won't spam the thread outside of my observations with a circuit that might be close.

[rccolgan]

It's alive! Sounds great. The 2k drive pot indeed works well. If unity is desired, maxing out the drive is necessary in this design. I need to play more with the recovery pot values as 1M goes from grounded signal to full within the first quarter turn. I'm also thinking 1m still might be necessary for proper bias and impedance. I'll need to research more on that.

My favorite feature might be the high headroom flat response recovery boost. With Drive and Recovery maxed out at 100% and boost at 0% (unity) I can tell there is an ever slight drop in level and some top end, which is every FX loop out there, right? . The boost recovers that drop and can add a hell of a punch at the top end of the sweep. I'd say the top end is smoothed out slightly instead of describing it as losing some clarity. Maybe this would be good and "by design" when used in an ODS FX loop.

¯⁠\⁠_⁠(⁠ツ⁠)⁠_⁠/⁠¯

[ijedouglas]

Great job! I may have to try one of these

[rccolgan]

Great job! I may have to try one o these

Thanks!! I need to fix the LED location on the faceplate and it should be ready for DIY scene soon once the revisions are made.

I forgot to add that I'll try an OPA2134PA op amp instead of the TL072 to see if that slight top-end loss is recovered. I know the bandwidth of the TL072 is well beyond audible range but I also like the rail-to-rail design of the OPA2134PA. It may also be the 150pf Silver Mica cap I used, a trick that overdrive circuits use in that spot to sweeten the top end. I have an NP0 MLCC cap to go in that place as an experiment as well.

[dbharris]

Looks great!

-Dan